Search

CN-122029613-A - Annotating medical images

CN122029613ACN 122029613 ACN122029613 ACN 122029613ACN-122029613-A

Abstract

Manual placement of markers in medical imaging is destructive to the workflow and prone to error. Accordingly, a computer-implemented method for annotating medical images is provided. The method comprises obtaining a medical image to be annotated and applying an annotation algorithm to the medical image. The annotation algorithm derives a entropy diagram from the obtained medical image to be annotated, identifies at least one region of the medical image that is a low entropy region within the entropy diagram, and determines values of one or more parameters to be applied to the acquired annotation of the medical image. The one or more parameters include image coordinates of a location of the post-acquisition annotation in the medical image, the image coordinates being within the at least one low entropy region. In addition, the method includes annotating the medical image by applying the post-acquisition annotation to the medical image using the determined values, and outputting the annotated medical image. The method improves workflow by automating marker placement.

Inventors

  • J. Kopenic
  • Trent B
  • A. Gosson
  • J. M. Mai
  • S. Long Ge
  • V. Stromer
  • N. Xinjinitz
  • KUEHL NORBERT

Assignees

  • 皇家飞利浦有限公司

Dates

Publication Date
20260512
Application Date
20241007
Priority Date
20231013

Claims (14)

  1. 1. A computer-implemented method for annotating a medical image, the method comprising: Obtaining a medical image (300) to be annotated; Applying an annotation algorithm to the medical image, wherein the annotation algorithm performs the following operations: a entropy diagram is derived from the acquired medical image (300) to be annotated, Identifying at least one region of the medical image (300), the at least one region being a low entropy region within the entropy diagram, Values of one or more parameters to be applied to a post-acquisition annotation (302) of the medical image are determined, Wherein the one or more parameters include image coordinates of a location of the post-acquisition annotation in the medical image, the image coordinates being within at least one of the low entropy regions, and Annotating the medical image by applying the post-acquisition annotation to the medical image using the determined values, and The annotated medical image is output.
  2. 2. The method of claim 1, wherein the one or more parameters further comprise a scaling factor for the post-acquisition annotation (302).
  3. 3. The method of claim 1 or 2, wherein the annotation algorithm determines the values of the one or more parameters based on the type of post-acquisition annotation (302).
  4. 4. A method according to claim 3, wherein the type of the post-acquisition annotation (302) corresponds to one or more of a sideways marker, a viewing position marker, an operator ID marker.
  5. 5. The method of any preceding claim, wherein the one or more parameters further comprise a brightness of the post-acquisition annotation (302), and wherein the annotation algorithm determines the brightness based at least in part on a background brightness of the medical image (300).
  6. 6. The method of any preceding claim, wherein the values determined by the annotation algorithm are stored in metadata of the medical image (300) together with the post-acquisition annotations (302).
  7. 7. The method of any of claims 1-5, wherein the annotation algorithm uses the determined values of the one or more parameters to draw the post-acquisition annotation (302) directly into the medical image (300).
  8. 8. The method of any preceding claim, wherein the annotation algorithm utilizes a trained machine learning model to determine the values of the one or more parameters.
  9. 9. The method of any preceding claim, wherein the annotation algorithm utilizes a model of an anatomical structure depicted in the medical image (300) to determine the image coordinates of the location of the post-acquisition annotation based at least in part on one or more landmarks included in the model.
  10. 10. The method of claim 9, wherein the annotation algorithm further utilizes a priori knowledge about the orientation and/or lateral aspect of the anatomical structure to determine the image coordinates.
  11. 11. The method of any preceding claim, wherein the annotation algorithm utilizes an image processing algorithm to select the values of the one or more parameters that avoid overlapping of the post-acquisition annotation (302) with at least one pre-identified region of the medical image (300) depicting diagnostically relevant structures and/or foreign objects.
  12. 12. The method of any preceding claim, wherein the annotation algorithm identifies at least one direct radiation zone of the medical image (300) based on entropy within the medical image, wherein the at least one direct radiation zone comprises suitable image coordinates of the location of the post-acquisition annotation (302).
  13. 13. A computing system (800) configured to perform the method of any preceding claim.
  14. 14. A computer-readable medium (804, 808) comprising instructions that, when executed by a computing system (800), cause the computing system to perform the method of any of claims 1-12.

Description

Annotating medical images Technical Field The invention relates to a method and a system for annotating medical images. Background In medical imaging, the anatomy of a patient can be placed in different orientations and can have different sidedness. For example, in a front-to-rear (AP) radiograph, the rear structure is closer to the detector, while in a rear-to-front (PA) image, the front structure is closer to the detector. Sidedness refers to which side of the body (i.e., left or right) is to be considered. A person may not be able to identify the orientation or lateral aspect of the patient's anatomy when viewing an X-ray image without prior knowledge. Thus, the image should be annotated and taken as part of the examination. Typically, a radiological technician manually annotates an image using pre-acquisition markers (e.g., guide markers indicating "l=left" or "r=right" placed in the field of view) or post-acquisition markers placed in the image with software. In either case, marker placement is destructive to the radiographic workflow and prone to error. Furthermore, marker placement may be subject to hospital-specific guidelines that govern the location and/or size of the markers. From US 2021/38365a 1a computer-implemented method of training a likelihood-based computational model for determining the position of an image representation of an annotated anatomical structure in a two-dimensional X-ray image is known. Disclosure of Invention To better address one or more of these concerns, in a first aspect of the invention, a computer-implemented method for annotating medical images is provided. The method includes obtaining a medical image to be annotated, and applying an annotation algorithm to the medical image, wherein the annotation algorithm derives a entropy diagram from the obtained medical image to be annotated, identifies at least one region of the medical image that is a low entropy region within the entropy diagram, and determines values of one or more parameters to be applied to a post-acquisition annotation of the medical image. The one or more parameters include image coordinates of a location of the post-acquisition annotation in the medical image, the image coordinates being within at least one of the low entropy regions. Additionally, the method includes annotating the medical image by applying the post-acquisition annotation to the medical image using the determined values, and outputting an annotated medical image. In contrast to the possibility approach disclosed in US2021/0383565A1, the annotation approach presented by the present invention allows the use of post-acquisition information. The one or more parameters may also include a scaling factor for the post-acquisition annotation. The one or more parameters may also include a brightness of the post-acquisition annotation. The one or more parameters may also include an orientation or rotation of the post-acquisition annotation, e.g., a horizontal orientation or a vertical orientation. The one or more parameters may also include a text direction of the annotation, e.g., left-to-right script or right-to-left script. The annotation algorithm may determine values of one or more parameters based on predetermined criteria. For example, the annotation algorithm may determine values of the one or more parameters based at least in part on the type of the post-acquisition annotation. The type of annotation may correspond to one or more of an off-side marker, a view position (orientation) marker, an operator ID marker. For example, the annotation algorithm may determine the size of the post-acquisition annotation based on the type and/or diagnostic relevance of the post-acquisition annotation, where, for example, the sideways or oriented indicia is greater than the operator ID indicia. The annotation algorithm may determine the values of the one or more parameters based at least in part on image content. For example, the annotation algorithm may determine the values of the one or more parameters to avoid overlapping of the post-acquisition annotations with at least one pre-identified region of the medical image depicting diagnostically relevant structures and/or foreign objects. In another example, the annotation algorithm may determine the values of the one or more parameters to place the post-acquisition annotation in at least one direct radiation zone of the medical image. Different priorities may be assigned to different criteria, wherein, for example, overlapping with diagnostic related structures assuming the highest priority is avoided. In the case of brightness as a parameter, the annotation algorithm may determine brightness of the post-acquisition annotation based at least in part on background brightness of the medical image. In this case, the predetermined criterion may be such that the brightness can be set such that the annotation is not too bright but still visually distinguishable from the background. In an example, by using